A large percentage of individuals globally are impacted by eye diseases that may lead to vision acuity if not detected early. The developments in artificial intelligence have greatly expanded the possibilities for early identification and accurate assessment of ocular disorders. This research work focused on the use of deep learning strategies to classify multiple eye disorders from OCT images. OCT imaging technique provides high-resolution and cross-sectional images of retinal layers that allow us to comprehend retinal disease and its structure in greater depth. Two benchmark data sets, OCTDL and OCTID, were used to classify relevant eye diseases such as AMD, DR, ERM, RAO, RVO, and VID. To enhance the robustness and accuracy of disease classification, the research incorporates data augmentation techniques such as flip, rotate, contrast, and zoom. This research work used advanced convolutional neural networks, including InceptionV3, Xception, and EfficientNetB0, coupled with the Adam optimizer and transfer learning techniques. The results indicated that the Inception model attained an accuracy of 96.91%, while the Xception model achieved an accuracy of 94.78%. Notably, EfficientNetB0 reached the highest validation accuracy at 97.12%.